About Me

I am Frank Shiwei Feng, a Machine Learning Engineer at  Apple, where I work on Neural Engine Compiler team to develop the software stack that enables fast and energy-efficient AI inference on dedicated hardware accelerators deployed across Apple’s product ecosystem.

I obtained my Ph.D. degree from the Department of Computer Science at Purdue University in Dec 2025. I am very fortunate to be advised by Samuel Conte Professor Xiangyu Zhang.

Previously, I obtained my B.Sc. degree in Computer Science as outstanding graduate from Nanjing University (NJU) in 2020. I am grateful to have Prof. Fengyuan Xu and Dr. Hao Wu as my initial advisors, who introduced me to the world of research. I also had the privilege to work with Dr. Lingfei Wu during my undergrad study.

My research interest lies in:

• Robust Performance of Autonomous Systems: software analysis for performance diagnostics in auto-driving, robotic and LLM agent systems.
• Safe Deployment of Foundation Models: focusing on robustness of self-supervised learning and multi-modal contrastive learning.

Working Experience

• 2026.01 - Present, Machine Learning Compiler Engineer, Apple, Sunnyvale, CA, USA

• 2025.05 - 2025.08, Machine Learning Compiler Intern, Apple, Sunnyvale, CA, USA

• 2024.05 - 2024.08, Machine Learning Compiler Intern, Apple, Sunnyvale, CA, USA
• 2020.07 - 2021.03, Machine Learning Compiler Engineer, Cambricon, Beijing, China

Education

• 2021.09 - 2025.12, Ph.D. in Computer Science, Purdue University, USA
• 2016.09 - 2020.06, B.Sc. in Computer Science, Nanjing University, China

Selected Publications (Full List)

Robust Autonomous Systems

• SpecOps: A Fully Automated AI Agent Testing Framework in Real-World GUI Environments

  Syed Yusuf Ahmed, Shiwei Feng, Chanwoo Bae, Calix Barrus, Xiangyu Zhang

  International Conference on Software Engineering (ICSE 2026)

• TAI3: Testing Agent Integrity in Interpreting User Intent

  Shiwei Feng, Xiangzhe Xu, Xuan Chen, Kaiyuan Zhang, Syed Yusuf Ahmed, Zian Su, Mingwei Zheng, Xiangyu Zhang

  Conference on Neural Information Processing Systems (NeurIPS 2025)

• Temporal Logic-Based Multi-Vehicle Backdoor Attacks against Offline RL Agents in End-to-end Autonomous Driving

  Xuan Chen, Shiwei Feng, Zikang Xiong, Shengwei An, Yunshu Mao, Guanhong Tao, Wenbo Guo, Xiangyu Zhang

  Conference on Neural Information Processing Systems (NeurIPS 2025)

• RFCAudit: AI Agent for Auditing Protocol Implementations Against RFC Specifications

  Mingwei Zheng, Chengpeng Wang, Xuwei Liu, Jinyao Guo, Shiwei Feng, Xiangyu Zhang

  IEEE/ACM International Conference on Automated Software Engineering (ASE 2025)

• EffiTune: Diagnosing and Mitigating Training Inefficiency for Parameter Tuner in Robot Navigation System

  Shiwei Feng, Xuan Chen, Zhiyuan Cheng, Zikang Xiong, Yifei Gao, Siyuan Cheng, Sayali Kate, Xiangyu Zhang

  IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2025)

• ROSCallBaX: Statically Detecting Inconsistencies In Callback Function Setup of Robotic Systems

  Sayali Kate, Yifei Gao, Shiwei Feng, Xiangyu Zhang

  Proceedings of the 2025 ACM SIGSOFT International Symposium on the Foundations of Software Engineering (FSE 2025)

  🎖 ACM SIGSOFT Distinguished Artifact Award

• ROCAS: Root Cause Analysis of Autonomous Driving Accidents via Cyber-Physical Co-mutation

  Shiwei Feng, Yapeng Ye, Qingkai Shi, Zhiyuan Cheng, Xiangzhe Xu, Siyuan Cheng, Hongjun Choi, Xiangyu Zhang

  IEEE/ACM International Conference on Automated Software Engineering (ASE 2024)

  🎖 ACM SIGSOFT Distinguished Paper Award

• Fusion is Not Enough: Single-Modal Attacks to Compromise Fusion Models in Autonomous Driving

  Zhiyuan Cheng, Hongjun Choi, Shiwei Feng, James Liang, Guanhong Tao, Dongfang Liu, Michael Zuzak, Xiangyu Zhang

  International Conference on Learning Representations (ICLR 2024)

• BadPart: Unified Black-box Adversarial Patch Attacks against Pixel-wise Regression Tasks

  Zhiyuan Cheng, Zhaoyi Liu, Tengda Guo, Shiwei Feng, Dongfang Liu, Mingjie Tang, Xiangyu Zhang

  International Conference on Machine Learning (ICML 2024)

Safety AI Deployment

• Distribution Preserving Backdoor Attack in Self-supervised Learning

  Guanhong Tao*, Zhenting Wang*, Shiwei Feng, Guangyu Shen, Shiqing Ma, Xiangyu Zhang

  Proceedings of the 45th IEEE Symposium on Security and Privacy (S&P 2024)

• LOTUS: Evasive and Resilient Backdoor Attacks through Sub-Partitioning

  Siyuan Cheng, Guanhong Tao, Yingqi Liu, Guangyu Shen, Shengwei An, Shiwei Feng, Xiangzhe Xu, Kaiyuan Zhang, Shiqing Ma, Xiangyu Zhang

  IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2024)

• Detecting Backdoors in Pre-trained Encoders [code]

  Shiwei Feng, Guanhong Tao, Siyuan Cheng, Guangyu Shen, Xiangzhe Xu, Yingqi Liu, Kaiyuan Zhang, Shiqing Ma, Xiangyu Zhang

  IEEE/CVF Conference on Computer Vision and Pattern Recognition 2023 (CVPR 2023)

• FLIP: A Provable Defense Framework for Backdoor Mitigation in Federated Learning [code]

  Kaiyuan Zhang, Guanhong Tao, Qiuling Xu, Siyuan Cheng, Shengwei An, Yingqi Liu, Shiwei Feng, Guangyu Shen, Pin-Yu Chen, Shiqing Ma, Xiangyu Zhang

  International Conference on Learning Representations 2023 (ICLR 2023)

  ECCV 2022 Workshop on Adversarial Robustness in the Real World (AROW 2022), Best Paper Award

• BEAGLE: Forensics of Deep Learning Backdoor Attack for Better Defense [code]

  Siyuan Cheng, Guanhong Tao, Yingqi Liu, Shengwei An, Xiangzhe Xu, Shiwei Feng, Guangyu Shen, Kaiyuan Zhang, Qiuling Xu, Shiqing Ma, Xiangyu Zhang

  Proceedings of the 30th Network and Distributed System Security Symposium (NDSS 2023)

• Graph-to-Tree Neural Networks for Learning Structured Input-Output Translation with Applications to Semantic Parsing and Math Word Problem [code]

  Shucheng Li*, Lingfei Wu*, Shiwei Feng, Fangli Xu, Fengyuan Xu and Sheng Zhong

  Conference on Empirical Methods in Natural Language Processing 2020. (EMNLP 2020 Findings)

Honors and Awards

• 2023-25 Travel Grants × 3 (NeurIPS 2025, ASE 2024, CVPR 2023).
• 2025.06 ACM SIGSOFT Distinguished Artifact Award at FSE 2025.
• 2025.03 Bounty Program about Novel LLM Agent Evaluation, supported by UK AI Security Institute (US$40,000).
• 2024.10 ACM SIGSOFT Distinguished Paper Award at ASE 2024.
• 2022.10 Best Paper Award at ECCV 2022 AROW Workshop.
• 2022.08 1st Place (tie) in AutoDriving CTF at DEF CON 30, top 4/110 teams, hosted by Baidu Security & UC Irvine.
• 2019.10 People’s First Class Scholarship, Ministry of Education, China.
• 2018.10 Suzhou Yucai Scholarship (10 recipients among 16 STEM departments), Nanjing University.
• 2017.09 National Scholarship (Top 2%), Ministry of Education, China.

Services

• Reviewer

  ICLR 2024-2026, ICML 2024-2026, CVPR 2025, ICCV 2025, NeurIPS 2023-2026, AAAI 2024-2026, ECCV 2024, MM 2024, IoT-J 2023-2024 (Journal)

• Artifact Evaluation Committee

  CCS 2023, ISSTA 2024, USENIX Security 2025

• Sub-reviewer

  ICSE-CAIN 2023, ASE 2023, FSE 2024-2025, ISSTA 2024, USENIX Security 2025

• Student Volunteer

  ICSE 2020

Last updated: Mar 30, 2026.